The Form 4 dataset provides structured, filing-level data from all Form 4 and Form 4/A filings submitted to the SEC's EDGAR system since January 2009. Each record represents one complete Form 4 filing and contains the reporting insider's identity and relationship to the company, the issuer's details, and all non-derivative and derivative transactions reported in that filing — including transaction dates, transaction codes, share quantities, prices, and post-transaction holdings. The dataset covers approximately 150,000 to 200,000 filings per year across roughly 5,000-6,000 domestic issuers and is distributed as compressed JSONL files (.jsonl.gz) organized by calendar month.
This is the primary data source for analyzing insider trading activity at US public companies — covering open-market purchases and sales, stock option exercises, equity compensation grants, gifts, and all other changes in beneficial ownership reported by corporate officers, directors, and 10% shareholders.
Programmatically retrieve the full list of dataset archive files, download URLs and dataset metadata.
Dataset Index JSON API
Download the entire dataset as a single archive file.
Download Entire Dataset:
Download a single container file (e.g. monthly archive) from the dataset.
Download Single Container:
Each JSONL record captures one Form 4 filing. The unit of observation is one filing, not one transaction. A filing may contain multiple transactions across multiple securities, all nested as arrays within the record.
Form 4 is the "Statement of Changes in Beneficial Ownership of Securities," filed under Section 16(a) of the Securities Exchange Act of 1934. It is the mechanism through which corporate insiders disclose every reportable change in their ownership of a company's equity securities.
| Field | Description | Example |
|---|---|---|
| accessionNo | Unique SEC accession number | "0001234567-25-000123" |
| formType | Original or amendment | "4" or "4/A" |
| filedAt | EDGAR acceptance datetime | "2025-03-15T16:30:00-04:00" |
| periodOfReport | Date of the earliest reported transaction | "2025-03-13" |
| Field | Description | Example |
|---|---|---|
| issuer.name | Company name | "Apple Inc" |
| issuer.cik | Issuer CIK | "0000320193" |
| issuer.tradingSymbol | Ticker symbol as reported by the filer | "AAPL" |
A filing can list multiple reporting owners (e.g., an individual and a trust they control). Each owner object contains:
| Field | Description | Example |
|---|---|---|
| name | Person or entity name | "Cook Timothy D" |
| cik | Owner CIK | "0001214156" |
| isDirector | Director flag | true / false |
| isOfficer | Officer flag | true / false |
| officerTitle | Title if officer | "CEO", "CFO", "SVP, General Counsel" |
| isTenPercentOwner | 10%+ owner flag | true / false |
| isOther | Other relationship flag | true / false |
Array of transaction objects for common stock and other non-derivative securities:
| Field | Description | Example / Values |
|---|---|---|
| securityTitle | Security name | "Common Stock" |
| transactionDate | Date of transaction | "2025-03-13" |
| transactionCode | SEC transaction code | P, S, A, M, G, F, C, J (see code table below) |
| sharesTraded | Shares transacted | 10000 |
| pricePerShare | Per-share price (null for gifts, some awards) | 178.50 |
| acquiredDisposedCode | Direction | "A" (acquired) or "D" (disposed) |
| sharesOwnedFollowingTransaction | Post-transaction holding | 150000 |
| directOrIndirectOwnership | Ownership type | "D" (direct) or "I" (indirect) |
| natureOfOwnership | Indirect holding vehicle | "By Spouse", "By Trust" |
| transactionTimeliness | Filing timeliness | blank (timely) or "L" (late) |
Array of transaction objects for options, warrants, convertible securities, and other derivatives:
| Field | Description | Example / Values |
|---|---|---|
| securityTitle | Derivative name | "Stock Option (Right to Buy)" |
| conversionOrExercisePrice | Strike or conversion price | 125.00 |
| transactionDate | Date of transaction | "2025-03-13" |
| transactionCode | SEC transaction code | Same codes as Table I |
| sharesTraded | Derivative units transacted | 5000 |
| pricePerShare | Price paid for the derivative | 0.00 |
| exerciseDate | Earliest exercisable date | "2026-03-13" |
| expirationDate | Expiration date | "2035-03-13" |
| underlyingSecurity.title | Underlying equity | "Common Stock" |
| underlyingSecurity.shares | Underlying share equivalents | 5000 |
| sharesOwnedFollowingTransaction | Remaining derivative holdings | 20000 |
| directOrIndirectOwnership | Ownership type | "D" or "I" |
Array of footnote objects, each with an id (e.g., "F1") and text string, cross-referenced by transaction rows. Footnotes explain vesting conditions, 10b5-1 plan adoption dates, charitable gift recipients, indirect ownership structures, and other qualifications.
| Code | Meaning | Analytical Signal |
|---|---|---|
| P | Open-market or private purchase | High — discretionary buying |
| S | Open-market or private sale | High — discretionary selling |
| A | Grant, award, or other acquisition | Medium — compensation event, not market view |
| M | Exercise or conversion of derivative | Medium — often paired with same-day S or F |
| F | Tax withholding via share disposition | Low — mechanical tax event |
| G | Bona fide gift | Low — estate/charitable planning |
| C | Conversion of derivative security | Low — contractual conversion |
| D | Disposition to the issuer | Context-dependent |
| J | Other acquisition or disposition | Context-dependent; check footnotes |
| W | Acquisition by will or succession | Low — involuntary |
| Z | Deposit into or withdrawal from voting trust | Low — structural |
Form 4 is filed by Section 16 reporting persons — insiders of companies with equity securities registered under Section 12 of the Exchange Act:
Officers — CEO, CFO, COO, principal accounting officer, any vice president in charge of a principal business unit, and other policy-making officers as defined by Rule 16a-1(f). Companies designate their Section 16 officers.
Directors — every board member of a Section 12 registrant, including independent directors and non-executive chairs.
Ten-percent beneficial owners — any person or entity owning more than 10% of any class of registered equity securities. This captures activist investors, founding shareholders, private equity sponsors post-IPO, and institutional holders crossing the threshold. Unlike officers and directors, 10% owners are subject to Section 16 only for the class in which they exceed 10%.
Form 4 covers issuers with at least one class of equity securities registered under Section 12:
Companies with only debt registered under Section 12 or reporting solely under Section 15(d) do not trigger Section 16 reporting.
Form 4 is event-driven, not periodic. Every reportable change in beneficial ownership triggers a filing obligation. Since the Sarbanes-Oxley Act of 2002, most transactions must be reported within two business days of the transaction date — making Form 4 a near-real-time data source for insider trading activity.
Reportable transactions include open-market purchases and sales (P, S), option exercises (M), stock grants and RSU vestings (A), tax withholding dispositions (F), gifts (G), conversions (C), and other changes in beneficial ownership.
Certain small or exempt transactions may be deferred to the annual Form 5, but in practice most insiders voluntarily report everything on Form 4.
Form 4 is the highest-volume individual filing type on EDGAR: approximately 150,000 to 200,000 filings per year. Volume clusters around post-earnings windows (when blackout periods lift), quarterly equity compensation vesting dates, year-end tax planning, and post-IPO lockup expirations.
| Authority | Provision |
|---|---|
| Section 16(a), Exchange Act | Mandates disclosure of insider ownership changes on Forms 3, 4, and 5 |
| Rules 16a-1 through 16a-13 | Define reporting persons, reportable transactions, and exemptions |
| Section 16(b), Exchange Act | Short-swing profit disgorgement for matched insider trades within six months |
| Section 403, Sarbanes-Oxley Act | Two-business-day filing deadline for Form 4 |
| Rule 10b5-1 | Governs pre-arranged trading plans; 2023 amendments added cooling-off periods, plan-adoption disclosure, and a Form 4 checkbox |
| Rule 144 | Governs resale of restricted and control securities; affiliates also file Form 4 |
Form 3 is a one-time holdings snapshot filed when a person becomes a Section 16 insider. It has no transaction tables, no transaction codes, and no price data. Form 4 reports every subsequent change with full transaction detail. Form 3 establishes a baseline; Form 4 is where the trading activity lives.
Form 5 is an annual catch-up filing for exempt or late transactions, due 45 days after fiscal year-end. It uses the same table structure as Form 4 but captures transactions that are predominantly small, exempt, or untimely. Form 5 volume is a small fraction of Form 4 volume and the annual delay eliminates its value for trading signal research.
Filed under Section 13(d)/13(g) when any person crosses 5% ownership. Key differences: different regulatory framework, lower threshold (5% vs. 10%), broader filer scope (any 5% holder, not just insiders), aggregate rather than transaction-level reporting, and less frequent amendments. Schedule 13D adds narrative about investment purpose and plans that Form 4 does not contain. Form 4 provides transaction-level granularity within two business days; 13D/13G provides aggregate ownership with 10-day (13D) or 45-day (13G) lag.
13F reports quarterly portfolio snapshots of institutional investment managers with $100M+ in 13(f) securities. The populations barely overlap: 13F covers institutional managers, Form 4 covers individual corporate insiders. 13F is quarterly with a 45-day lag and shows position sizes without transaction detail. Form 4 is within two business days and shows individual transactions with dates, prices, and types.
Form 144 is a prospective notice filed before selling restricted or control securities under Rule 144. Form 4 is the retrospective report of the actual completed transaction. The same insider files both, but Form 144 shows intended maximum shares while Form 4 shows actual shares transacted.
This dataset contains the same information as raw EDGAR XML filings but in a flat, analysis-ready JSONL format. Raw filings require XML parsing, multi-document joins, and handling of schema variations across filing years. This dataset delivers each filing as a single structured JSON record.
Quantitative portfolio managers use insider transaction signals as alpha factors in systematic equity selection models. They aggregate net insider buying per issuer over rolling windows, filter to informative transaction codes (P and S), weight by officer seniority, and combine the result with other factors. Cluster buying detection — flagging issuers where multiple insiders buy simultaneously — is a well-documented predictive signal.
Equity research analysts monitor insider activity for conviction signals. Large open-market purchases by C-suite executives during share price weakness, unusual selling patterns outside 10b5-1 plans, and changes in post-transaction ownership levels all feed into fundamental analysis and stock recommendations.
Securities litigation attorneys rely on Form 4 data as evidence in 10b-5 class actions (insider selling as scienter evidence), Section 16(b) short-swing profit recovery (mechanical matching of purchases and sales within six months), and shareholder derivative suits challenging executive compensation.
Corporate governance professionals at proxy advisory firms and institutional investors track insider alignment, screen for short-swing violations, verify compliance with director stock ownership guidelines, and monitor 10b5-1 plan usage patterns.
SEC enforcement staff and exchange surveillance teams cross-reference insider transactions against material event timelines to identify suspicious pre-announcement trading, monitor Section 16(a) filing timeliness, and build longitudinal trading profiles for pattern analysis.
Compliance officers at public companies verify that all Section 16 insiders file timely and accurately, cross-check filings against blackout period calendars and pre-clearance records, and compile Section 16(a) delinquency disclosures for the annual proxy statement.
Financial journalists source breaking stories from notable insider transactions, report on aggregate insider sentiment trends, and contextualize corporate events by checking whether executives traded beforehand.
Academic researchers use Form 4 data to study the predictive power of insider trades, information asymmetry around corporate events, equity compensation structures, and the effectiveness of regulatory changes such as the Sarbanes-Oxley filing deadline and the 2023 Rule 10b5-1 amendments.
Quantitative funds build tradable insider sentiment factors by aggregating open-market purchases (code P) and sales (code S) per issuer over rolling windows. They filter out mechanical transactions (codes A, M, F, G), weight by officer seniority and transaction size relative to holdings, exclude 10b5-1 plan trades, and combine the signal with momentum, value, and earnings-revision factors. Form 4 is the only data source providing transaction-level insider trades with two-business-day timeliness, officer titles, and transaction type codes.
Plaintiff-side securities attorneys systematically match all Form 4 purchases and sales by the same insider within rolling six-month windows to identify short-swing profits recoverable under Section 16(b). The matching algorithm applies the lowest-in, highest-out method (Smolowe v. Delendo Corp.) to maximize computed profit. This screening runs across thousands of issuers using the structured transaction arrays and is entirely dependent on Form 4 data.
Compensation consultants and proxy advisory analysts track the lifecycle of equity compensation: initial grants (code A in Table II), vesting events (code A in Table I or M for exercises), tax withholdings (code F), and open-market sales (code S). Summing pricePerShare times sharesTraded across all dispositions yields the total value realized by each executive per year — a figure that complements the proxy statement's target compensation disclosures.
Regulators cross-reference Form 4 transaction dates against material events (earnings, M&A, FDA decisions, restatements) to flag cases where insiders traded in the 30-60 days before disclosure in a direction consistent with the undisclosed news. The two-business-day filing deadline and the transactionDate field (as distinct from the filing date) enable precise temporal alignment.
Institutional stewardship teams build longitudinal insider ownership profiles by aggregating sharesOwnedFollowingTransaction across all Form 4 filings per insider-issuer pair. They track whether C-suite executives and directors are net buyers or net sellers over multi-year periods, flag companies with declining insider ownership, and use this data in engagement conversations and proxy voting decisions on say-on-pay and director elections.
Fundamental analysts and special-situations managers screen for issuers where three or more insiders make open-market purchases (code P) within a 30-day window with no insider selling. Filtering for meaningful transaction sizes (over $25,000) and prioritizing companies where the CEO or CFO is among the buyers produces a high-conviction watchlist that academic literature has shown carries predictive value.
Governance researchers and investigative journalists analyze the 10b5-1 plan checkbox and footnote disclosures to identify patterns in pre-arranged trading plan usage — plans adopted shortly before positive news, terminated before negative news, or modified frequently. This analysis evaluates the effectiveness of the SEC's 2023 10b5-1 amendments and flags potential abuse.
Transaction codes drive analytical value. The most informative codes are P (purchase) and S (sale), which reflect discretionary decisions. Code M (exercise) often pairs with a same-day Table I disposition (S or F), representing option exercise and immediate sale or tax withholding. Researchers constructing insider sentiment signals typically filter to P and S only.
Price per share is null or zero for many transaction types. Grants (A), gifts (G), and RSU vestings often report $0.00 or null. Only open-market purchases (P) and sales (S) reliably carry market prices.
Amendments (4/A) restate the entire filing. De-duplicate by keeping only the latest filing date for a given accession number or issuer-owner-period combination.
Multiple reporting owners per filing are common when a trust, LLC, or family member holds the securities and the controlling individual is also a reporting person. The same underlying transaction may appear on separate Form 4 filings by different reporting owners.
Indirect ownership rows indicate beneficial ownership through trusts, spouses, children, or partnerships. The natureOfOwnership field and footnotes explain the holding structure.
10b5-1 plan indicator: Since the 2023 Rule 10b5-1 amendments, Form 4 includes a checkbox indicating whether the transaction was executed under a pre-arranged plan. Combined with footnote disclosures about plan adoption dates, this enables systematic separation of pre-planned from discretionary trades.
The full dataset is available as a ZIP archive containing monthly .jsonl.gz files. Each line in a JSONL file is one Form 4 filing record in JSON format. Files are organized by calendar year and month (e.g., 2026/2026-03.jsonl.gz). Coverage begins January 2009 and updates daily as new filings are processed from EDGAR.
Download: https://api.sec-api.io/datasets/form-4.zip
Form types included: 4, 4/A
Total archive size: ~910 MB compressed
What is the difference between Form 4 and Form 4/A? Form 4 is the original filing. Form 4/A is an amendment that replaces the entire filing. If both exist for the same reporting event, use the amendment (most recent filing date) and discard the original.
Why does one transaction sometimes appear on multiple Form 4 filings? When beneficial ownership is shared — for example, an executive controls a trust that holds shares — both the individual and the trust may file separate Form 4s reporting the same transaction. De-duplication requires matching on issuer, transaction date, security, and share count.
Why is pricePerShare null or zero on many records? Non-market transactions such as grants (code A), gifts (code G), RSU vestings, and some option exercises might not report a price or report $0.00.
How do I identify the most analytically significant transactions? Filter to transaction codes P (purchase) and S (sale), which represent discretionary open-market trading. Exclude codes A, M, F, and G, which represent compensation events, mechanical exercises, tax withholdings, and gifts. Further filter out transactions flagged as 10b5-1 plan trades if you want to isolate discretionary insider views.
Does the dataset include the insider's total holdings? Each transaction row includes a sharesOwnedFollowingTransaction field showing the insider's beneficial ownership after that transaction. By tracking this field across sequential filings, you can reconstruct an insider's ownership trajectory over time.
How does this dataset relate to the SEC's EDGAR full-text filings? This dataset contains the same information as the raw XML Form 4 filings on EDGAR, extracted and structured into analysis-ready JSONL format. No content is added or removed — the dataset is a structured representation of the original filing.